Higher wages or lower expectations? - IABdoku.iab.de/discussionpapers/2015/dp0615.pdf · Higher wages or lower expectations? ... the vacancy and the firm itself. ... They reason that

IAB Discussion PaperArticles on labour market issues

6/2015

Hanna BrenzelAnne Müller

ISSN 2195-2663

Higher wages or lower expectations?Adjustments of German firms in the hiring process

Higher wages or lower expectations? Adjustments of German firms in the hiring process

Hanna Brenzel (IAB) Anne Mueller (IAB)

Mit der Reihe „IAB-Discussion Paper“ will das Forschungsinstitut der Bundesagentur für Arbeit den Dialog mit der externen Wissenschaft intensivieren. Durch die rasche Verbreitung von Forschungsergebnissen über das Internet soll noch vor Drucklegung Kritik angeregt und Qualität gesichert werden.

The “IAB-Discussion Paper” is published by the research institute of the German Federal Employment Agency in order to intensify the dialogue with the scientific community. The prompt publication of the latest research results via the internet intends to stimulate criticism and to ensure research quality at an early stage before printing.

IAB-Discussion Paper 6/2015 2

Contents

Abstract .................................................................................................................... 4

Zusammenfassung ................................................................................................... 4

1 Introduction .......................................................................................................... 5

2 Theoretical background and related literature ...................................................... 6

3 Data and assumptions ......................................................................................... 8

4 Descriptive analysis ........................................................................................... 11

5 Empirical results ................................................................................................ 14

6 Conclusion ......................................................................................................... 21

References ............................................................................................................. 22

Appendix ................................................................................................................ 24


Abstract

Labour shortages are a field of research that has been investigated quite thoroughly. The reactions of firms facing problems during the hiring process are, however, large-ly neglected in empirical literature. Our research will fill this empirical gap and shed light on the question of whether reactions according to the neoclassical theory or to the Reder Hypothesis are more common in reality.

We make use of a unique dataset, the German Job Vacancy Survey, which allows us to observe the entire operational recruitment process including potential prob-lems, concessions made by firms as well as characteristics of the hired candidate, the vacancy and the firm itself.

Whether concessions are made mainly depends on the labour market situation and on the specific hiring problems of a firm. We also find that firms are rather flexible in their reactions in accordance with the specific hiring problem. Therefore, both theo-ries seem to apply in reality.

Zusammenfassung

Das Thema Arbeitskräfteengpässe wurde bereits umfangreich erforscht, allerdings blieben dabei die Reaktionen von Betrieben mit Schwierigkeiten im Einstellungspro-zess weitestgehend unberücksichtigt. Unsere Forschung soll diese Lücke schließen und Aufschluss darüber geben ob auf dem deutschen Arbeitsmarkt eher die ne-oklassische Theorie oder aber die Reder-Hypothese Anwendung findet.

Für unsere Analysen greifen wir auf die IAB-Stellenerhebung zurück. Dieser Daten-satz ermöglicht es den gesamten Rekrutierungsprozess zu beobachten, einschließ-lich der Probleme die dabei auftreten können, mögliche Zugeständnisse seitens der Unternehmen als auch personen-, stellen- und firmenspezifische Charakteristika.

Ob Zugeständnisse gemacht werden hängt vor allem von der Arbeitsmarktlage und den spezifischen Einstellungsproblemen eines Unternehmens ab. Die Ergebnisse zeigen auch, dass die Unternehmen entsprechend den spezifischen Einstellungs-problemen flexibel reagieren. Daraus lässt sich ableiten dass beide Theorien An-wendung auf dem deutschen Arbeitsmarkt finden.

JEL classification: J33, J01, D22

Keywords: Hiring process, vacancy, wage and qualification adjustments, estab-lishment survey


1 Introduction After the financial crises in the late 2000s and the following economic boom in Ger-many, research fields such as labour shortages and related labour market adjust-ments and their corresponding determinants and consequences have spurred a strong interest in Germany. One driving force can be seen in the increased labour demand and the fact that the number of vacancies has almost returned to pre-crisis levels. Moreover, the looming demographic change and its impact on the labour market is heating up the labour shortages debate. Despite the fact that the effects of the demographic change are not discernible yet all over the German labour market, they stir up fear amongst the population, the firms and politics. So, there is a risk of facing labour shortages and in particular skill shortages in the near future. We can already observe a sharp increase in the number of firms facing problems during the hiring process. In 2013, the share of these firms compared to all firms with at least one new hire reached over 33 per cent, which implies an increase of about 10 per-centage points during the last seven years.1

These developments have led a number of researchers to investigate both the de-terminants and the consequences of difficulties during the hiring process and in par-ticular of the alleged skill shortages. Nevertheless, empirical studies largely neglect-ed the firms’ responses. There are, however, different theories on how firms can react when they face difficulties in hiring new employees. We focus on two of them: the neoclassical approach – firms pay higher wages – and the Reder Hypothesis – firms lower their demands regarding skills. This paper 1) tries to fill the research gap that exists when it comes to firms’ reactions and 2) sheds light on the question which theory is more supported by the empirical results. We make use of a unique dataset, the German Job Vacancy Survey, which allows us to examine whether firms make concessions while hiring new employees. We control for difficulties with-in the hiring process, the firms’ characteristics as well as attributes of the respective vacancy and the applicants’ characteristics. Our aim in this article is not to explain the alleged skill shortages or to restrict the problems during the hiring process to skill shortages only. In fact, analyzing what determines concessions in general as well as the type of concession is the main contribution of this paper.

The main challenge will be to expose the determinants which may affect the firms’ reactions or to detect that there are no specific characteristics which determine the different concessions firms make.

The paper is organized as follows: Initially, a brief overview of the underlying theory and the related literature is given. From these, a series of hypotheses will be derived and embedded in the theoretical framework of the investigation. Subsequently, the data used will be presented and discussed. The fourth part analyses the results of the binary and the multinomial logit model. The paper closes with a conclusion.

1 See Appendix Figure 1.


2 Theoretical background and related literature Our work is essentially based on the well-known traditional neoclassical theory and an extension of this theory, the so called “Reder Hypothesis”. The neoclassical theo-ry implies the existence of a perfect labour market with the inherent characteristic of balancing supply and demand: if the labour demand exceeds the labour supply, wages will increase until a new labour market equilibrium is reached and demand and supply of labour are balanced again. So the adjustment of wages can be seen as an equalization mechanism in order to balance the differences between supply and demand. The Reder Theory is an extension of this theory and opens up an al-ternative way to handle the discrepancies of the labour market. The idea behind this approach is that firms not only adjust wages and take qualification and skills for granted, but also use qualification and skills as an equalization mechanism.

Established in 1955, the hypothesis states that in times of high-skilled workers shortage firms try to fill their vacancies by promoting medium-ability workers (Lud-steck and Haupt 2007). That means employers tend to reduce their pre-established minimum requirements in order to find new employees. Thus, wage bargaining power of the higher skilled will be reduced. But this mechanism only operates if eve-ry vacancy on the labour market can be filled with a worker. Ludsteck and Haupt (2007) and Büttner et al. (2010) review the Reder Hypothesis empirically for Germa-ny. The authors conclude that lower quantiles of the wage distribution respond more sensitively to labour demand changes than the upper part of the wage distribution. Büttner et al. (2010) reach the same conclusion: the occupational wage composition is, as the Reder Hypothesis assumes, pro-cyclical. In times of recessions, the skill level of new hires rises significantly within the occupational structure and decreases in times of upturns. However, the effect is much lower in small firms than in large firms.

As mentioned in the introduction, there are a number of studies which deal with re-lated issues but rarely with the operational responses that can be observed when a vacancy was difficult to fill. The following sections outline thematically related litera-ture. Nevertheless, these findings can only provide clues about determinants that may be influential since they focus on difficulties in the search and hiring process and on the duration of the search but not on operational reactions. Furthermore, the literature stock differs vastly in its research design.

One stock of literature deals with the determinants of an expected labour shortage (Haskel and Martin 1993; Green et al. 1998; Haskel and Martin 2001; Watson et al. 2006). Building on that, others analyse influencing factors on the search duration and recruitment problems associated with resulting adjustment mechanisms (Brencic 2009; Stevens 2007; Andrews et al. 2008; Baldwin et al. 1995; Gorter and Van Ommeren 1999). Baldwin et al. (1995), for example, used descriptive analysis and found out that recruitment problems mainly occur due to a lack of suitable appli-cants, but other problems such as wages and working conditions also play a central role.


Watson’s et al. (2006) conclusion implies that the firms’ size is a significant determi-nant in skills shortages perception: the larger the firm the greater the expectation to face hiring problems. One reason can be seen in the context of more foresighted human resources planning of larger firms compared to smaller ones. Furthermore, Watson et al. (2006) concludes that growing firms have a higher skill shortage per-ception. Haskel and Martin (2001) distinguish between skill shortages, hard-to-fill vacancies, and hiring difficulties. They reason that firms operating in the high-tech sector complain about a skills shortage in particular, whereas firms which use mi-croelectronics in word processing announce problems or difficulties in the process of hiring. Arrow et al. (1959) analysed the set of problems in terms of required skilled labour from a differentiated perspective. In doing so, they developed a model which explains the dynamics of the market adjustment process in order to analyse the sci-entist-engineer “shortage” in the US in the 1950s. Their research is based on the neoclassical theory approach which leads them to the conclusion that the interaction between the rising demand of engineers and the lagging adjustment concerning the wages might be a feasible explanation for what happened at that time. Wieling and Borghans (2001) refer to the neoclassical approach, too, and argue that the dis-crepancies between supply and demand in segments of the labour market can oc-casionally be balanced by wage adjustments, but also adjustments via flexible work-ing hours and lower qualification requirements can be witnessed. The paper by Fang (2009) is of great importance for our study. He examines the responses of Canadian firms to difficulties during the hiring process by using linear probability models and probit models. He concludes that employers rather tend to consider short-term and less costly solutions. However, he finds no evidence for wage in-creases or voluntary fringe benefits. The actual reactions which may be used by companies with staffing problems remain unclear in almost all studies.

Based on this overview we can state that labour shortages and their reception are caused by various factors. If they occur, different reactions are imaginable: for ex-ample professional training programs, adoption of flexible working hours, increase in overtime hours, implementation of self-directed work groups, adjustments in the requirements for qualification or work experience, and as the neoclassical approach suggests, wage concessions (Fang 2009).

In this paper we concentrate on two of these possibilities – wage concession and adjustment of the required qualification or work experience of the applicant. We fo-cus on these reactions because compared to the other possibilities, adjustment in wages and requirements are easily observable and present the most widespread reactions to solve insufficient labour supply. Furthermore, the two adjustments could also appear simultaneously, as has been emphasized in the classical contribution by Reder (1955). Identifying potential determinants for the different reactions will be our effort in the following.


3 Data and assumptions Our investigation is based on the German Job Vacancy Survey (JVS) of the Institute for Employment Research (Kettner et al. 2011). This written cross-sectional estab-lishment survey has been conducted every fourth quarter of a year since 1989. It allows representative statements about the total number of vacancies for East and West Germany according to the size of firms and 23 economic branches. As such it is unparalleled for the German labour market in this regard. The JVS also contains detailed information about the recruitment process of firms.2 We use data from the fourth quarter of 2011 and 2012, when a set of questions treating the wage setting process was introduced. In total, over 15,000 establishments participated each year. Almost 10,000 of them had at least one new hire in the past 12 months and hence represent our sample of interest.

In the questionnaire, firms answered questions about the most recent case of a new hire. They were asked whether wage concession and/or adjustments of qualification and work experience took place in this case. For our analysis we use two different questions about remuneration or adjustments within the hiring process. The first corresponding question in the survey is: “Does the employee differ in qualification, work experience, age or other important aspects from what you primarily searched for?” First of all, the respondents had to answer “Yes” or “No”. In a second step, the firms were asked: “If yes, how does she/he differ?” Possible answers were: “Qualifi-cation is higher or lower” and “Experience is greater or smaller”.3 We recoded these answers and generated a binary variable which is one if the firm indicated that either the minimum qualification or the experience of the newly hired employee was lower than primarily scheduled. This variable shall reflect the already mentioned Reder Hypothesis.

Despite the issue of work or experience adjustments, we are also interested in knowing whether a firm had to pay more in order to fill a certain vacancy. The sec-ond question of interest in the survey is therefore: “Was higher payment required in order to fill this vacancy?” Again, the firms had two choices of answers: “Yes” or “No”.

Our empirical analysis is divided into two main steps: First, a multivariate logit model is used to find influencing factors on the probability to make concessions in the first place. In a second step, we use a multinomial model to find out which variables ex-plain whether concessions are made via wage adjustments, adjustments of qualifi-

2 Further information about the JVS can be found at http://www.iab.de/en/befragungen/stellenangebot.aspx. The JVS data are available for non-commercial empirical research via the Data Research Centre of the Federal Employment Agency: http://fdz.iab.de/en.aspx.

3 Also, the options “Employee is younger or older” or “Other differences” were given. Note that the last item offers an open answer which was given in 30 cases. We defined these cases as missing.


http://www.iab.de/en/befragungen/stellenangebot.aspx

http://fdz.iab.de/en.aspx

cation and/or experience or both. The following paragraphs present possible influ-encing factors regarding vacancy-specific, person-specific, firm-specific, as well as labour market characteristics and their assumed effect.4

For the first step, a binary variable was generated which is 1 if either qualification or experience concessions were made or if the firm had to raise the salary and 0 if no adjustments at all were made.

We assume that concessions are more likely if it was hard to fill a vacancy. The JVS includes a question concerning possible problems that occurred during the hiring process. The information derived from this question represents our measure for a difficult staffing process and should be one main determinant for adjustments made. We assume that the existence of problems leads to a higher probability to make adjustments in order to successfully fill the vacancy.

As vacancy-specific characteristics we use information about the required minimum qualification of the new employee, about further special requirements such as man-agerial skills, the possibility to negotiate the payment, and information about the type of the contract. We expect a positive influence on concessions for an increase of the required qualification as well as for the possibility of salary negotiation driven by an increase in wage adjustments. If a temporary contract exists firms might have to make more concessions because the vacancy is less attractive. The same should be true for part-time jobs.

The age of the new employee, the gender and his or her prior employment status reflect the person-specific characteristics in our model. For these variables we can only tell whether they correlate with concessions or not, but we cannot prove wheth-er they influence the probability of concessions. The reason is that the characteris-tics of the hired new employee can either be the reason for the concessions that were made but can also be the result of problems and the corresponding conces-sions. Nevertheless, we control for these variables since we cannot exclude that they might influence the outcome. While it is hard to make assumptions on whether existing correlations will be positive or negative, we certainly will interpret the re-sults.

Finally, we include various firm-specific characteristics in our models. One of the strongest arguments against observing concessions can be seen in the collective bargaining coverage of a firm, which may hinder individual wage adjustments. We included the existence of collective bargaining agreements as a dummy variable. We also included the firms’ expected employment trend for the near future as an explanatory variable. It is imaginable that an increase in the expected number of

4 Please note that our sample contains only successful recruitment processes, even if they might have been difficult. We therefore know who was finally hired and can include char-acteristics of this new employee into our models.


employees will influence concessions in a positive way. The expected growth in a firm’s workforce could increase potential skills shortages and increase the necessity to make concessions in order to attract applicants.

Furthermore, characteristics of the firms’ staff are included, such as the proportion of employed minors and the proportion of temporary workers in a firm. Our expectation is that a high share of minor employment or temporary employment – which is on average cheaper than permanent, full-time employment – is connected to more con-cessions since vacancies in such a firm might be less attractive. Also, we include the share of female staff. Since women still often earn less than men, one can as-sume that a higher share of women in a firm may correlate with a lower average payment and therefore more possibilities to make wage concessions if necessary.

The size of the firm, measured by the number of employees, functions as a proxy for the financial scope of the firm giving it more or less scope to pay higher wages. Al-so, the branch of industry of a firm and its location in East or West Germany as well as the regional unemployment rate are taken into account. We expect that a higher regional unemployment rate leads to fewer concessions because the large supply of potential new employees makes it easier for firms to recruit.

In order to account for the interactions between the two different adjustments – wage or qualification/experience – a second step of analysis is conducted. The dependent variable in this case is measured on a nominal scale, that is only wage adjustments, only qualification and/or experience adjustments, or both types of ad-justments simultaneously. On the one hand, this enables us to identify the influential determinants on the particular responses of the firms, and on the other hand, allows statements in which cases the firms tend to react via wage concessions or via ad-justments respecting qualification and experience, or if the firms use both channels of adjustment.

One of the most important independent variables in our model again can be seen in the difficulty to fill the vacancy. To find out whether the firms’ reactions depend on the type of hiring difficulty they face, we not only include the presence of hiring diffi-culties in our second model, but rather take the different reasons for these difficulties into account. The firms were asked whether they had problems filling the respective vacancy and if so, what caused these problems. The questionnaire asked about four different reasons:5 an inadequate occupational qualification of the applicants, an insufficient number of applicants, a lack of willingness to accept the standard work-ing conditions, and the fact that the applicants required too high wages (or any com-bination of these causes).

5 An overview of the distribution of the different adjustments according to the reasons of difficulties is given in Appendix Table 1.


To examine further determinants and factors that play a significant role in the deci-sion-making process of a firm, we again include vacancy-specific, person-specific and firm-specific variables described above in this second step.

4 Descriptive analysis A total of 9.4 million vacancies were filled externally in Germany in 2011 and 2012. This corresponds to around 1.4 million firms. Around 24 per cent of the firms which had at least one new hire in 2011 or 2012 react via wage concession or lowering the required qualification or work experience. If we distinguish between the different possibilities of concessions, we find that ten per cent of the firms react with paying higher wages. In 12 per cent, the adjustments made concerned work experience or qualification and only around two per cent of firms responded with both types of concession. These findings are highlighted in Table 1.

Table 1 The frequencies and weighted shares of the different adjustments

Frequency Weighted Weighted

share (in %) No concessions 11,979 1,057,700 76 Only wage concessions 1,072 134,719 10 Only qualification/experience adjustments (lower) 1,299 163,213 12 Both type of concession 257 34,031 2

Source: German Job Vacancy Survey 2011 and 2012. The survey weights vary across 23 economic sectors and 7 firm size classes. Weighted share is conducted according to the firms with at least one new hire in 2011 or 2012.

Table 2 shows some descriptive evidence concerning the existence of concessions within a firm considering vacancy-specific, person-specific and firm-specific charac-teristics which might determine the decision of a firm to react using at least one of the three possible mechanisms. In order to take into account the sampling frame, we weight the share of concessions. In line with our expectations, 48 per cent of the firms facing difficulties to fill the vacancy react by adopting at least one of the three possible mechanisms. Surprisingly, more than 13 per cent of firms which do not re-port difficulties during the hiring process also report concessions. Regarding the vacancy-specific characteristics we find a clear connection between the required skills and the concessions firms make: The higher the minimum required qualifica-tion, the higher the share of concessions. Around 27 per cent of firms which de-manded at least a higher qualification for the vacancy conceded. If we take other requirements such as special skills and knowledge into account, we also observe higher shares of concessions. For example, if a firm requires managerial skills or long-term experience, concessions were made in around 29 respectively 27 per cent compared to 23 respectively 20 per cent if those skills are not required. Of course, if a firm is willing or able to negotiate the wage, the share of concession rises from 16 to 36 per cent. This might mostly be driven by wage rather than qualification and/or experience adjustments. Other vacancy-specific characteristics concerning the con-tract such as temporal dimensions or working hours show quite different results.


According to our descriptive analysis, it does not matter whether the contract is tem-porary or not, whereas it does matter whether the job is a part-time job.

If we switch to person-specific characteristics, we observe no noteworthy differences between male and female new employees, but we do find differences according to the previous employment status of the hired person. The share of concessions is rather small if the person was self-employed or in apprenticeship in the correspond-ing firm compared to persons who were employed elsewhere, who were unem-ployed or out of labour force. On the face of it, this may seem a bit peculiar, but we have to remember that the options for adjustments are quite diverse. On the one hand, a high number of concessions made in case of previously unemployed per-sons can be driven by the adjustments via lower qualification or experience whereas a high number of concessions can be attributed to wage concession when previous-ly employed persons are concerned.

Surprisingly, if a firm is bound by a collective wage agreement, the share of conces-sions does not differ much. This could be put down to difficulties paying a higher wage than the union wage. So, the only mechanism available for these firms is to lower qualification or experience. Firms not bound by collective wage agreements also have the option to react by making wage concessions. In the end, the probabil-ity to react by implementing at least one of the possible mechanisms should be more or less the same.

Firms with an expected increase in their workforce show the highest share of con-cessions. An expected employment increase should imply more frequent search processes and hence the firms’ risk of facing skills shortages and hiring difficulties might also increase. Furthermore, also firms with an expected decrease in their workforce make more concessions than firms with an expected stable number of employees. This might be due to a lack of attractiveness of shrinking firms, making concessions necessary to find applicants.

If we distinguish between five firm sizes, we can identify the smallest firms with less than ten employees as those with the highest share of concessions. With an as-cending employment stock the share of concessions decreases. This can be ex-plained by the fact that large firms rather tend to offer higher wages right from the beginning. Smaller firms may start with a lower wage offer, making it possible that they have to adjust their wages more often in order to find a new employee. The frequency of concessions across industries ranges from eight per cent in public ad-ministration to 29 per cent in firms belonging to the construction sector.

Finally, on the regional level we do not find extensive differences between the shares of concessions. Summing up, it seems that vacancy-specific as well as per-son-specific and firm-specific characteristics correlate with the share of concessions.


Table 2 Concession according to firm-specific characteristics

Share of conces-sions (in %)

No concession

Conces-sion raw weighted

Difficulties to fill the vacancy

Yes 2,143 1,511 41 48 No 9,836 1,117 10 13

Minimum required qualification

Unqualified (without occupational training) 1,433 200 12 20

Intermediate qualification 8,468 1,897 18 24 Higher qualification 2,078 531 20 27

Managerial skills No 10,896 2,280 17 23 Yes 1,083 348 24 29

Long-term experience in occupational field

No 5,912 1,040 15 20 Yes 6,067 1,588 21 27

Wage negotiation No 8,880 1,200 12 16 Yes 3,099 1,428 32 36

Fixed-term employment contract

No 6,562 1,585 19 24 Yes 5,417 1,043 16 24

Part-time job No 8,400 2,065 20 27 Yes 3,579 563 14 17

Female No 6,154 1,511 20 25 Yes 5,825 1,117 16 22

Previous employment status

Employed outside the firm 6,039 1,266 17 24 Unemployed 3,228 883 21 26 Temporary agency worker in this firm 356 40 10 23

Self-employed 405 79 16 16 Apprentice in this firm 358 24 6 11 Elsewhere in apprenticeship/ education 1,136 273 19 23

Out of labour force (housewife etc.) 457 63 12 24 Collective wage agree-ment

No 6,256 1,687 21 25 Yes 5,723 941 14 22

Expected employment trend…

increase in employees 2,854 912 24 33 no change 7,807 1,435 16 20 decrease in employees 1,318 281 18 29

Firm size

<10 1,410 396 22 26 10 - 49 5,901 1,426 19 24 50 - 249 3,316 596 15 19 250 - 500 665 121 15 14 500+ 687 89 11 14

Economic sectors

Agriculture, hunting and forestry 362 79 18 25 Mining and quarrying, electricity, gas and water supply, recycling 1,014 158 13 18

Manufacturing 2,356 632 21 24 Construction 452 116 20 29 Trade, maintenance and repair of motor vehicles, motor 451 115 20 24

Transportation and storage 508 126 20 27 Information and communication 444 130 23 21 Financial intermediation, insurance 410 67 14 17 Industry-related services 1,355 369 21 28 Other services 3,614 755 17 21 Public administration 1,013 81 7 8

Region West 6,566 1,398 18 24 East 5,413 1,230 19 22

Total 11,979 2,628 18 24 Source: German Job Vacancy Survey 2011 and 2012. The survey firm-weights vary across 23 economic sectors

and 7 firm size classes. Regression sample (see Table 3) with 14,607 observations.


5 Empirical results In order to verify whether the identified determinants also influence the probability of adjustments in a multivariate analysis and under control for other variables, we es-timated a binary logit model with concession/no concession as the dependent varia-ble.6

As Table 3 shows, the econometric results are in line with our previous descriptive evidence. If firms report difficulties in the hiring process, the probability of adjust-ments via one of the three possibilities increases by almost 20 percentage points. We also find a significant but small increasing effect in terms of the required mini-mum qualification. If the firm requires a higher qualification instead of no qualifica-tion, the probability to observe concession increases by around two percentage points. The same holds true for special skills and knowledge requirements: if the vacancy requires managerial skills or long-term experience in an occupational field, the probability to concede increases by 1.3 respectively 1.4 percentage points. As we have already seen in the descriptive part, the probability to concede increases by around ten percentage points if the firm is able or willing to negotiate the wage. However, if the contract is fixed-term or if the job is part-time, the effect is rather small but points in the same direction as the descriptive results.

Person-specific characteristics such as the gender variable or the age of the person who was hired do not influence the probability of adjustments. In contrast, the previ-ous employment status does show significant results although the effects are not big. However, we need to be careful in interpreting these results. The problem here is that it is impossible to identify the direction of causality. To give an example: It is possible for a firm to lower its skills requirements because the specific person who is to be hired was previously unemployed and lacks some skills. Simultaneously, it could be that the firm was not able to find any applicants and consequently had to lower the requirements and only subsequently found a previously unemployed per-son who was finally hired.

Firms bound by a collective wage agreement are significantly less likely to make any concessions although the effect here is really small: the probability decreases by not even one percentage point. Also, the workforce composition does not play a major role in terms of adjustments. We only find a small however significant effect for the share of female employees. The higher the share of female employees, the higher the probability to concede is. As we have already seen in Table 2, establishments with an expected increase or decrease in their workforce are more likely to react by using one of the adjustment mechanisms compared to firms expecting a stable number of employees. As before, the effect is a bigger if the firm expects an in-crease instead of a decrease. According to the findings concerning firm size, we find a significant and negative effect with an ascending employment stock. In other

6 The estimation is based on unweighted data. The key results do not change when esti-mating a probit or complementary log log model.


words, smaller firms are significantly more likely to concede and the probability de-creases with an increase in firm size. Of course, one might argue that bigger firms are in a better position to pay higher wages. Therefore, they would have a smaller probability to face difficulties and hence do not have to make wage concessions at all. But since our model controls for difficulties during the hiring process, we only compare firms of different sizes but facing exactly the same problems or difficulties.

Compared to the manufacturing sector, establishments in most branches show a lower probability to make adjustments. The largest effect can be seen for public ad-ministration which is government-financed and often uses very strict and detailed job descriptions.

We also find a highly significant relationship between the regional unemployment rate and the probability of concessions. If the regional unemployment rate increases by one per cent, the probability for concessions decreases by 2.1 per cent. This might be due to the strengthened bargaining power of employers when unemploy-ment rates are high.

Finally, we see a significant effect depending on whether the firm is located in East or West Germany. According to our results, firms located in East Germany have a one per cent higher probability to concede, independently of the difficulties they have during the hiring process. Overall, we can conclude that all types of character-istics – vacancy-specific, person-specific and firm-specific – influence the probability of concession. However, difficulties to fill the vacancy along with external circum-stances such as tight labour markets are the main driving forces, of course.


Table 3 Probability of concessions - Marginal effects based on a logit model dy/dx Standard error Difficulties to fill the vacancy (dummy) 0.191 *** 0.019 Minimum required qualification (dummy)

Unqualified (without occupational training) (reference) --- --- Intermediate qualification 0.016 *** 0,005 Higher qualification 0.018 *** 0,006

Special skills and knowledge (dummy) Managerial skills 0.013 ** 0.006 Long-term experience in occupational field 0.014 *** 0.004

Wage negotiation (dummy) 0.092 *** 0.012 Fixed-term employment contract (dummy) 0.006 * 0.003 Part-time job (dummy) -0.008 ** 0.003 Female (dummy) -0.005 0.003 Age 0.000 * 0.000 Previous employment status (dummy)

Employed outside the firm (reference) --- --- Unemployed 0.026 *** 0.004 Temporary agency worker in this firm -0.016 * 0.008 Self-employed -0.001 0.008 Apprentice in this firm -0.023 *** 0.007 Elsewhere in apprenticeship/education 0.018 *** 0.006 Out of labour force (housewife etc.) 0.000 0.009

Collective wage agreement (dummy) -0.007 ** 0.003 Share of female employees 0.016 * 0.008 Share of part-time employees -0.009 0.007 Share of temporary employees -0.014 0.009 Expected employment trend…(dummy)

increase in employees 0.022 *** 0.006 no change (reference) --- --- decrease in employees 0.017 ** 0.007

Plant size (dummy) < 10 (reference) --- --- 10 - 49 -0.010 ** 0.004 50 - 199 -0.016 *** 0.004 200 - 499 -0.017 *** 0.006 500+ -0.028 *** 0.006

Economic branches (dummy) Agriculture, hunting and forestry -0.005 0.008 Mining and quarrying, electricity, gas and water supply, recycling -0.012 ** 0.005 Manufacturing (reference) --- --- Construction 0.000 0.007 Trade, maintenance and repair of motor vehicles, motor -0.003 0.008 Transportation and storage -0.001 0.007 Information and communication -0.013 ** 0.006 Financial intermediation, insurance -0.013 * 0.007 Industry-related services -0.003 0.006 Other services -0.006 * 0.005 Public administration -0.021 *** 0.006

East Germany (dummy) 0.009 ** 0.004 Regional unemployment rate -0.213 *** 0.056 Year (dummy)

Year 2011 -0.004 0.003 Year 2012 (reference)

McFadden's Adj. R2 0.171 Area under ROC curve 0.793

Source: German Job Vacancy Survey 2011 and 2012. 14,607 observations; regional unemployment rates are for 418 administrative districts (Landkreise und kreisfreie Städte – NUTS3 regions); Standard errors clustered at regional level. Marginal effects calculated for the reference group where dummies are ze-ro and the values of the shares and the unemployment rate are at the mean. The probability of negoti-ation for the reference group is 0.179.


After examining the determinants which influence the probability of any of the abovementioned concessions, we are now interested in what influences the firms’ choice of concession. For this purpose, we make use of a multinomial logit model and restrict our sample to those firms which report at least one of the concession possibilities.7 The dependent variable now includes three different categories: Only wage concession, only adjustment in qualification or experience, and both types of concessions. This enables us to set all three possible reactions of a firm in relation to each other. In order to gain a clear picture, we distinguish between the different causes of hiring difficulties. Table 4 shows that if the reason is “inadequate occupa-tional qualification of the applicants”, the probability to observe only wage conces-sions decreases by about 12 per cent, while the probability of both types of conces-sion is not affected and the probability of only qualification/experience adjustment increases by about 12 per cent. So, if the labour supply is inadequately qualified, firms tend to lower their requirements and hire a person who does not completely conform to what the firm wanted.

On the other hand, if the difficulty to fill the vacancy is caused by too high wage de-mands of the applicants, the probability to adjust the requirement regarding qualifi-cation or work experience declines by around 17 per cent. At the same time, this type of hiring difficulties increases the probability to observe only wage concessions by almost ten per cent and both types of concessions by seven per cent. Here, in line with the previous results, firms also show a reaction which can be linked directly to the particular reason of hiring difficulties. If the problem is seen in the lack of will-ingness to accept the standard working conditions, neither the probability of wage concession nor the probability of qualification or experience adjustments change significantly. An interesting issue arises in cases where an insufficient number of applicants is the reason for problems. Here we cannot anticipate ex ante which con-cessions the firms might make. As Table 4 shows, the probability for wage conces-sions only decreases to a small five per cent and the probability for qualification or experience adjustments increases by almost the same percentage. If only an insuffi-cient number of applicants is available, firms seem to anticipate that paying more alone might not lead to any results. On the other hand, reducing the requirements can attract a different group of potential workers who did not respond to the job offer before.

7 Summary statistics of the multinomial regression sample are provided in Appendix Ta-ble 2.


Table 4 Results of the multinomial logit model Only wage concession Qualification and/or Experience is

lower Both types of concession

Marginal effects Standard error Marginal effects Standard error Marginal effects Standard error Difficulties to fill the vacancy (dummy) Inadequate occupational qualification of the applicants -0.121 *** 0.029 0.119 *** 0.030 0.002 0.008

Too high wage demand 0.101 *** 0.034 -0.170 *** 0.037 0.069 ** 0.030 Lack of willingness to accept the standard working

conditions 0.011 0.024 -0.020 0.028 0.009 0.011

Insufficient number of applicants -0.046 ** 0.020 0.042 ** 0.021 0.004 0.008 Minimum required qualification (dummy) Unqualified (without occupational training) (reference) --- --- --- --- --- ---

Intermediate qualification 0.018 0.036 -0.010 0.039 -0.008 0.013 Higher qualification 0.048 0.042 -0.028 0.046 -0.020 0.015

Special skills and knowledge (dummy) Managerial skills 0.087 ** 0.037 -0.122 *** 0.041 0.035 * 0.020 Long-term experience in occupational field 0.047 ** 0.021 -0.056 ** 0.022 0.008 0.008

Wage negotiation (dummy) 0.385 *** 0.035 -0.465 *** 0.021 0.080 *** 0.030 Fixed-term employment contract (dummy) -0.087 *** 0.024 0.105 *** 0.026 -0.018 ** 0.009 Part-time job (dummy) -0.017 0.023 0.034 0.024 -0.017 ** 0.010 Female (dummy) 0.009 0.021 -0.021 0.025 0.012 0.010 Age 0.005 *** 0.001 -0.004 *** 0.001 0.000 0.000 Previous employment status (dummy) Employed outside the firm (reference) --- --- --- --- --- ---

Unemployed -0.150 *** 0.035 0.160 *** 0.036 -0.010 0.008 Temporary agency worker in this firm -0.087 * 0.050 0.083 0.059 0.004 0.026 Self-employed -0.098 *** 0.038 0.107 ** 0.046 -0.009 0.017 Apprentice in this firm -0.018 0.068 0.025 0.080 -0.008 0.031 Elsewhere in apprenticeship/education -0.125 *** 0.034 0.130 *** 0.036 -0.004 0.012 Out of labour force (housewife etc.) -0.137 *** 0.038 0.137 *** 0.045 0.000 0.021

Collective wage agreement (dummy) -0.033 * 0.018 0.048 ** 0.021 -0.015 * 0.009 Share of female employees 0.010 0.042 -0.022 0.047 0.012 0.019 Share of part-time employees -0.064 0.050 0.117 * 0.060 -0.053 ** 0.025 Share of temporary employees 0.161 ** 0.074 -0.194 ** 0.075 0.032 0.030 Expected employment trend…(dummy) increase in employees -0.012 0.018 0.002 0.020 0.010 0.010

no change (reference) --- --- --- --- --- --- decrease in employees -0.003 0.028 -0.026 0.037 0.029 0.021


Only wage concession Qualification and/or Experience is lower Both types of concession

Marginal effects Standard error Marginal effects Standard error Marginal effects Standard error Plant size (dummy) < 10 (reference) --- --- --- --- --- ---

10 - 49 0.060 ** 0.029 -0.061 * 0.033 0.001 0.009 50 - 199 0.073 ** 0.029 -0.072 ** 0.034 -0.001 0.011 200 - 499 0.045 0.045 -0.035 0.046 -0.010 0.017 500+ 0.136 ** 0.064 -0.128 * 0.066 -0.008 0.019

Economic branches (dummy) Agriculture, hunting and forestry 0.051 0.054 -0.026 0.057 -0.025 0.017 Mining and quarrying, electricity, gas and water supply, recycling 0.025 0.041 -0.002 0.044 -0.022 * 0.013

Manufacturing (reference) --- --- --- --- --- --- Construction -0.057 0.041 0.088 ** 0.043 -0.031 ** 0.015 Trade, maintenance and repair of motor vehicles, motor 0.067 0.048 -0.059 0.053 -0.008 0.015 Transportation and storage 0.108 ** 0.043 -0.118 ** 0.048 0.010 0.018 Information and communication 0.173 *** 0.060 -0.173 *** 0.064 0.000 0.015 Financial intermediation, insurance 0.124 * 0.070 -0.154 * 0.083 0.030 0.034 Industry-related services 0.101 ** 0.043 -0.108 ** 0.045 0.007 0.012 Other services 0.084 ** 0.034 -0.087 ** 0.038 0.003 0.012 Public administration -0.065 0.054 0.085 0.057 -0.020 0.018

East Germany (dummy) 0.063 ** 0.029 -0.062 ** 0.031 -0.002 0.010 Regional unemployment rate -0.780 ** 0.362 0.901 ** 0.377 -0.121 0.137 Year (dummy) Year 2011 0.011 0.016 -0.009 0.018 -0.002 0.007

Year 2012 (reference) --- --- --- --- --- --- N=2,643 y = Pr(outcome = 1) = 0.221 y = Pr(outcome = 1) = 0.732 y = Pr(outcome = 1) = 0.047 LR chi2 = 1.252*** McFadden's Adjusted R2 = 0.198 Maximum Likelihood R2 = 0.377 Count R2 = 0.695

Source: German Job Vacancy Survey 2011 and 2012. 2,643 observations; regional unemployment rates are for 418 administrative districts (Landkreise und kreisfreie Städte – NUTS3 regions). Standard errors clustered at regional level. Marginal effects calculated for the reference group where dummies are zero and the values of the shares and the unemployment rate are at the mean.


In contrast to Table 4, the minimum required qualification does not affect the firm’s choice significantly while special requirements such as managerial skills and long-term occupational experience raise the probability of wage concession by nine re-spectively five per cent. Meanwhile, the probability to observe concessions concern-ing qualification or work experience declines quite strongly by 12 per cent if mana-gerial skills are required and by six per cent if experience is required. The effect on both types of concession simultaneously varies between the two coefficients. Re-quested managerial skills increase the probability of making both types of conces-sions, whereas long-term experience does not affect the probability of both conces-sions. These results do not surprise since the variables suggest that the firm was looking for a special type of qualification. In such case, lowering requirements con-cerning skills or experience as far as necessary to fill the vacancy does not seem to be the appropriate reaction.

Of course, if the potential new employee has the opportunity to negotiate the wage, the probability to face wage concessions or both types of concessions increases significantly. If possible, applicants use their bargaining power and therefore the probability to observe higher wage increases while qualification or experience ad-justments become more unlikely. Contrariwise, if the bargaining power of the new employee is low because he or she is only needed for a limited period of time, firms rather react by adjusting qualification or experience requirements than by paying more. If the job is part-time, we do not find any significant effect on wage or qualifi-cation/experience adjustments. Only both types of concession simultaneously are slightly affected.

Person-specific characteristics play a significant role in the decision which conces-sions a firm makes, too. Again, we have to keep in mind that we cannot say any-thing about the direction of causality; the effects only reflect a correlation between the adjustments and covariates. According to the results, the older a person, the higher the probability of wage concessions is and the lower the probability of qualifi-cation or experience adjustments. If the hired person was previously unemployed, self-employed, in apprenticeship or somehow out of the labour force, lower require-ments regarding qualification or work experience occurred more often, but higher wages decreased compared to a hired person who had been employed previously. Again, we cannot verify whether the characteristics of the hired person influence the firms’ concessions or whether the concessions lead to hiring the respective person.

Firms bound by a collective wage agreement have a lower probability of making wage concession only or both concessions simultaneously but a higher probability of making qualification or experience adjustments. Our deliberations from chapter 4 apply here as well: paying salaries above the union wage is rather difficult and therefore firms have to respond by making qualification and experience adjustments.

The share of part-time employees and the share of temporary workers in the firm have a significant impact on the choice of concession. While a higher share of part-


time employment leads to a higher probability of qualification or experience adjust-ments and lowers the probability for a combination of both concessions, a high share of temporary employment increases the probability for higher wages and de-creases the probability of qualification or experience adjustments. In contrast to our findings concerning the basic decision to make concessions in the first place, the firms’ expected employment trend does not have any effect on the decision which concession to make. As we already found in our binary logit model, firm size plays a significant role in the decision process. The bigger the firm, the higher the probability is to make wage-related concessions, whereas the probability of qualification or ex-perience adjustments only decreases. Only medium-sized firms with 200 to 499 em-ployees do not differ significantly from small firms. However, our assumption that small firms with fewer than ten employees have less scope for pay raises and there-fore reduce their qualification requirements more frequently can be confirmed alto-gether.

Also, the branch of industry influences the type of concession. In Transportation and storage, Information and communication, Financial intermediation and insurance, Industry-related services, and Other services the probability to make wage conces-sions is higher than in the manufacturing sector. At the same time, in all branches mentioned above, the probability for qualification or experience adjustments is lower than in Manufacturing. For firms in the tertiary sector it hence seems to be easier to adjust wages. Finally, also the external circumstances of a firm play a role: In East-ern Germany, firms rather raise the wage than lower their requirements. However, the effect is small and should not be over-interpreted. In contrast, the regional un-employment rate has a noteworthy influence: the higher the rate, the lower the bar-gaining power of an applicant and the lower the probability that a firm will adjust the wage. Simultaneously, as the unemployment rate increases, it becomes more likely that firms lower their requirements concerning qualification or work experience.

Our findings point to the fact that firms seem to be relatively flexible in their reactions to possible problems and that they adjust their strategy accordingly with respect to the corresponding hiring situation.

6 Conclusion Our paper examines the determinants and characteristics that influence wage or qualification and work experience adjustments in the hiring process from a firm’s perspective. This empirical analysis is based on the German Job Vacancy Survey. For the first time, it is conducted with a focus on wage concessions and qualification or work experience adjustments connected with difficulties during the hiring process. The overall findings support the assumption that difficulties to fill the vacancy along with external circumstances such as tight labour markets are the main driving forces when it comes to making concessions in the first place.

Our empirical results also point out that firms are rather flexible in choosing the ap-propriate reaction depending on the specific situation. Accordingly, firms are, for


example, likely to adjust their qualification or experience requirements if the appli-cants are inadequately qualified and tend to adjust the wage if hiring problems occur due to too high wage demands. Besides the hiring situation itself, characteristics of the vacancy (such as required managerial skills), of the hired person (such as the previous employment status), and of the firm (such as firm size) are linked to the different types of concessions.

All things considered, we found evidence for the Reder Hypothesis as well as for the neoclassical approach of dealing with hiring difficulties. Although we cannot say which theory is the “correct” one, concessions implied by the Reder Hypothesis – such as adjustments in qualification or experience – are much more probable than the neoclassical wage concessions. Furthermore, observing both concessions sim-ultaneously is rather unlikely.

Skilled labour has recently become scarcer on the German labour market. One rea-son can be seen in the demographic change the country faces. The observation that firms concede in accordance with their specific hiring situation is a strong indicator that efficient market forces exist, making government intervention unnecessary. Nevertheless, especially small firms have to make a disproportionately high number of concessions. In light of the increasing labour market tightness it is arguable how long they will be able to remain successful in the competition for skilled labour.

References Andrews, M. J.; Bradley. S.; Stott, D. & Upward, R. (2008): Successful Employer Search? An Empirical Analysis of Vacancy Duration Using Micro Data. Economica, 75(299), 455–480.

Arrow, K. J. & Capron, W. M. (1959): Dynamic Shortages and Price Rises: The En-gineer-Scientist Case. The Quarterly Journal of Economics, 73(2), 292–308.

Baldwin, S.; Campbell, M. & Froud, M. (1995): Skill shortages and recruitment diffi-culties in Yorkshire and Humberside. The Regional Review.

Brencic, V. (2009): Employers' hiring practices, employment protection, and costly search: A vacancy-level analysis. Labour Economics, 16(5), 461–479.

Büttner, T., Jacobebbinghaus. P. & Ludsteck, J. (2010): Occupational Upgrading and the Business Cycle in West Germany. Economics: The Open-Access. Open-Assessment E-Journal, 4(10), 1–37.

Fang, T. (2009): Workplace responses to vacancies and skill shortages in Canada. International Journal of Manpower, 30(4), 326–348.

Gorter, C. & Van Ommeren, J. (1999): Sequencing. timing and filling rates of re-cruitment channels. Applied Economics, 31(10), 1149–1160.

Green, F.; Machin, S. & Wilkinson, D. (1998): The Meaning and Determinants of Skill Shortages. Oxford Bulletin of Economics and Statistics, 60(2), 165–187.

Haskel, J. & Martin, C. (2001): Technology, wages, and skill shortages: evidence from UK micro data. Oxford Economic Papers, 53(4), 642–658.


Haskel, J. & Martin, C. (1993): The Causes of Skill Shortages in Britain. Oxford Economic Papers, 45(4), 573–588.

Kettner, A.; Heckmann, M.; Rebien, M.; Pausch, S.; Szameitat, J. (2011): Die IAB-Erhebung des gesamtwirtschaftlichen Stellenangebots – Inhalte. Daten und Metho-den. In: Zeitschrift für Arbeitsmarkt Forschung, Jg. 44. H. 3. S. 245–260.

Ludsteck, J. & Haupt, H. (2007): An Empirical Test of the Reder Hypothesis. Mün-chener Wirtschaftswissenschaftliche Beiträge (VWL) 2007–11.

Reder, M. W. (1955): The Theory of Occupational Wage Differentials. The American Economic Review, 45(5), 833–852.

Stevens, P. A. (2007): Skill shortages and firms' employment behaviour. Labour Economics, 14, 231–249.

Watson, D.; Johnson, S. & Webb, R. (2006): Employer perceptions of skills deficien-cies in the UK labour market: a subregional analysis. Environment and Planning A, 38(9), 1753–1771.

Wieling, M. & Borghans, L. (2001): Descrepancies between Supply and Demand and Adjustment Processes in the Labour Market. LABOUR, 15(1), 33–56.


Appendix

Appendix Figure 1 Share of firms with problems during the hiring process as percentage of all firms with at least one new hire in the listed year

Source: German Job Vacancy Survey

Appendix Table 1 The weighted shares of the different adjustments according to the hiring difficulties

Reasons for difficulties No concession

Only wage concession

Qualification and/or Experi-ence is lower

Both types of concession

Inadequate occupational qualifi-cation of the applicants 47 13 31 9

Too high wage demand 41 23 25 12 Lack of willingness to accept the standard working conditions 52 14 26 8

Insufficient number of applicants 50 16 26 8 Source: German Job Vacancy Survey 2011 and 2012. The survey weights vary across 23 economic sectors and

7 firm size classes. Weighted share is conducted according to the firms with at least one new hire in 2011 or 2012.


Appendix Table 2 Summary statistics of the multinomial regression sample Variables Mean Std. Dev. Dependent multinomial variable 1.693 0.641 Difficulties to fill the vacancy (dummy)

Inadequate occupational qualification of the applicants 0.389 0.488 Too high wage demand 0.212 0.409 Lack of willingness to accept the standard working conditions 0.189 0.392 Insufficient number of applicants 0.362 0.481

Minimum required qualification (dummy) Unqualified (without occupational training) 0.076 0.265 Intermediate qualification 0.722 0.448 Higher qualification 0.202 0.401

Special skills and knowledge (dummy) Managerial skills 0.132 0.339 Long-term experience in occupational field 0.603 0.489

Collective wage agreement (dummy) 0.359 0.480 Wage negotiation (dummy) 0.543 0.498 Fixed-term employment contract (dummy) 0.397 0.489 Part-time job (dummy) 0.215 0.411 Female (dummy) 0.426 0.495 Age 35.707 10.556 Previous employment status (dummy)

Employed outside the firm 0.482 0.500 Unemployed 0.335 0.472 Temporary agency worker in this firm 0.015 0.122 Self-employed 0.030 0.171 Apprentice in this firm 0.009 0.095 Elsewhere in apprenticeship/education 0.104 0.306 Out of labour force (housewife etc.) 0.024 0.154

Expected employment trend…(dummy) increase in employees 0.347 0.476 no change 0.546 0.498 decrease in employees 0.107 0.310

Share of female employees 0.427 0.291 Share of part-time employees 0.183 0.227 Share of temporary employees 0.082 0.146 Plant size (dummy)

< 10 0.150 0.357 10 - 49 0.543 0.498 50 - 199 0.227 0.419 200 - 499 0.047 0.211 500 + 0.034 0.181

Economic branches (dummy) Agriculture, hunting and forestry 0.030 0.170 Mining and quarrying, electricity, gas and water supply, recycling 0.060 0.237 Manufacturing 0.241 0.428 Construction 0.044 0.205 Trade, maintenance and repair of motor vehicles, motor 0.045 0.207 Transportation and storage 0.048 0.213 Information and communication 0.049 0.216 Financial intermediation, insurance 0.026 0.158 Industry-related services 0.140 0.347 Other services 0.288 0.453 Public administration 0.031 0.172

East Germany (dummy) 0.468 0.499 Regional unemployment rate 0.080 0.034

Source: German Job Vacancy Survey 2011 and 2012. 2,643 observations; “unqualified” means without occupa-tional training; “higher qualification” indicates college or university degree.


Recently published

No. Author(s) Title Date 20/2014 Kruppe, T.

Lang, J. Labour market effects of retraining for the unem-ployment

8/14

21/2014 Klinger, S. Weber, E.

On GDP-Employment Decoupling in Germany 8/14

22/2014 Horbach, J. Determinants of labor shortage - with particular focus on the German environmental sector

8/14

23/2014 Doerr, A. Fitzenberger, B. Kruppe, T. Paul, M. Strittmatter, A.

Employment and earnings effects of awarding training vouchers in Germany

9/14

24/2014 Capuano, S. Hauptmann, A. Schmerer, H.-J.

Trade and unions: Can exporters benefit from collective bargaining?

10/14

25/2014 Homrighausen, P.

Differential pricing and private provider perfor-mance

11/14

26/2014 Fuchs, M. Weyh, A.

Demography and unemployment in East Germa-ny: How close are the ties?

12/14

27/2014 Stephan, G. Uthmann, S.

Akzeptanz von Vergeltungsmaßnahmen am Ar-beitsplatz: Befunde aus einer quasi-experimen-tellen Untersuchung published als: University of Lüneburg. Working paper series in economics : 321

12/14

28/2014 Wapler, R. Werner, D. Wolf, K.

Active labour-market policies in Germany: Do regional labour markets benefit?

12/14

29/2014 Dahmann, S. Anger, S.

The impact of education on personality: Evidence from a German high school reform

12/14

1/2015 Moczall, A. The effect of hiring subsidies on regular wages 1/15 2/2015 Stops, M. Revisiting German labour market reform effects:

A panel data analysis for occupational labour markets

1/15

3/2015 Brunow, S. Grünwald, L.

Exports, agglomeration and workforce diversity: An empirical assessment for German Establish-ments

1/15

4/2015 Osiander, C. Dietz, M.

What could all the money do? Ergebnisse eines faktoriellen Surveys zur Bedeutung von Oppor-tunitätskosten bei Weiterbildungsentscheidun-gen

1/15

5/2015 Lucht, M. Haas, A.

The productivity effect of migrants: Wage cost advantages and heterogeneous firms

2/15

As per: 2015-02-09

For a full list, consult the IAB website http://www.iab.de/de/publikationen/discussionpaper.aspx


http://www.iab.de/183/section.aspx/Publikation/k140825301




http://www.iab.de/183/section.aspx/Publikation/k141017j01



http://www.iab.de/183/section.aspx/Publikation/k141215j01










http://www.iab.de/de/publikationen/discussionpaper.aspx

Imprint

IAB-Discussion Paper 6/2015

Editorial addressInstitute for Employment Research of the Federal Employment AgencyRegensburger Str. 104D-90478 Nuremberg

Editorial staffRegina Stoll, Jutta Palm-Nowak

Technical completionGertrud Steele

All rights reservedReproduction and distribution in any form, also in parts, requires the permission of IAB Nuremberg

Websitehttp://www.iab.de

Download of this Discussion Paperhttp://doku.iab.de/discussionpapers/2015/dp0615.pdf

ISSN 2195-2663 For further inquiries contact the authors:

Hanna BrenzelPhone +49.911.179 5940E-mail [email protected]

Anne MüllerPhone +49.911.179 6767E-mail [email protected]

Documents

Higher wages or lower expectations? - IABdoku.iab.de/discussionpapers/2015/dp0615.pdf · Higher wages or lower expectations? ... the vacancy and the firm itself. ... They reason that